episode: 0 training return: tensor(-499.4158, device='cuda:0', grad_fn=<AddBackward0>)
episode: 1 training return: tensor(-499.9531, device='cuda:0', grad_fn=<AddBackward0>)
episode: 2 training return: tensor(-499.8977, device='cuda:0', grad_fn=<AddBackward0>)
episode: 3 training return: tensor(-494.0224, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 1 test_true_pfm: -418.85397678524487
episode: 4 training return: tensor(-499.9710, device='cuda:0', grad_fn=<AddBackward0>)
episode: 5 training return: tensor(-499.7875, device='cuda:0', grad_fn=<AddBackward0>)
episode: 6 training return: tensor(-499.9675, device='cuda:0', grad_fn=<AddBackward0>)
episode: 7 training return: tensor(-499.9270, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 2 test_true_pfm: 1293.495088564774
episode: 8 training return: tensor(-499.9464, device='cuda:0', grad_fn=<AddBackward0>)
episode: 9 training return: tensor(-499.9572, device='cuda:0', grad_fn=<AddBackward0>)
episode: 10 training return: tensor(-499.5651, device='cuda:0', grad_fn=<AddBackward0>)
episode: 11 training return: tensor(-499.9437, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 3 test_true_pfm: 50.94009695965721
episode: 12 training return: tensor(-499.9681, device='cuda:0', grad_fn=<AddBackward0>)
episode: 13 training return: tensor(-499.9023, device='cuda:0', grad_fn=<AddBackward0>)
episode: 14 training return: tensor(-499.2780, device='cuda:0', grad_fn=<AddBackward0>)
episode: 15 training return: tensor(-499.9572, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 4 test_true_pfm: 4051.272474326261
episode: 16 training return: tensor(-498.5963, device='cuda:0', grad_fn=<AddBackward0>)
episode: 17 training return: tensor(-499.9410, device='cuda:0', grad_fn=<AddBackward0>)
episode: 18 training return: tensor(-499.9666, device='cuda:0', grad_fn=<AddBackward0>)
episode: 19 training return: tensor(-499.9661, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 5 test_true_pfm: 393.57842249163053
episode: 20 training return: tensor(-499.9678, device='cuda:0', grad_fn=<AddBackward0>)
episode: 21 training return: tensor(-499.9641, device='cuda:0', grad_fn=<AddBackward0>)
episode: 22 training return: tensor(-499.9664, device='cuda:0', grad_fn=<AddBackward0>)
episode: 23 training return: tensor(-499.8884, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 6 test_true_pfm: 2747.659289957612
episode: 24 training return: tensor(-499.9436, device='cuda:0', grad_fn=<AddBackward0>)
episode: 25 training return: tensor(-499.9655, device='cuda:0', grad_fn=<AddBackward0>)
episode: 26 training return: tensor(-499.9683, device='cuda:0', grad_fn=<AddBackward0>)
episode: 27 training return: tensor(-493.0130, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 7 test_true_pfm: -358.2544953701649
episode: 28 training return: tensor(-467.6460, device='cuda:0', grad_fn=<AddBackward0>)
episode: 29 training return: tensor(-499.9476, device='cuda:0', grad_fn=<AddBackward0>)
episode: 30 training return: tensor(-499.7146, device='cuda:0', grad_fn=<AddBackward0>)
episode: 31 training return: tensor(-499.8449, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 8 test_true_pfm: 1253.2116429122264
episode: 32 training return: tensor(-447.2626, device='cuda:0', grad_fn=<AddBackward0>)
episode: 33 training return: tensor(-499.9547, device='cuda:0', grad_fn=<AddBackward0>)
episode: 34 training return: tensor(-499.9330, device='cuda:0', grad_fn=<AddBackward0>)
episode: 35 training return: tensor(-490.7505, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 9 test_true_pfm: 1806.7207786326783
episode: 36 training return: tensor(-499.9671, device='cuda:0', grad_fn=<AddBackward0>)
episode: 37 training return: tensor(-499.9383, device='cuda:0', grad_fn=<AddBackward0>)
episode: 38 training return: tensor(-499.6838, device='cuda:0', grad_fn=<AddBackward0>)
episode: 39 training return: tensor(-499.9686, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 10 test_true_pfm: 38.10708629637054
episode: 40 training return: tensor(-499.9662, device='cuda:0', grad_fn=<AddBackward0>)
episode: 41 training return: tensor(-499.9214, device='cuda:0', grad_fn=<AddBackward0>)
episode: 42 training return: tensor(-499.9665, device='cuda:0', grad_fn=<AddBackward0>)
episode: 43 training return: tensor(-499.9700, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 11 test_true_pfm: 489.0346529883332
episode: 44 training return: tensor(-499.9695, device='cuda:0', grad_fn=<AddBackward0>)
episode: 45 training return: tensor(-499.9642, device='cuda:0', grad_fn=<AddBackward0>)
episode: 46 training return: tensor(-494.1707, device='cuda:0', grad_fn=<AddBackward0>)
episode: 47 training return: tensor(-499.9693, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 12 test_true_pfm: 2864.315062307361
episode: 48 training return: tensor(-499.9417, device='cuda:0', grad_fn=<AddBackward0>)
episode: 49 training return: tensor(-499.8696, device='cuda:0', grad_fn=<AddBackward0>)
episode: 50 training return: tensor(-448.0346, device='cuda:0', grad_fn=<AddBackward0>)
episode: 51 training return: tensor(-499.8985, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 13 test_true_pfm: 1900.8720742127643
episode: 52 training return: tensor(-499.9680, device='cuda:0', grad_fn=<AddBackward0>)
episode: 53 training return: tensor(-499.9702, device='cuda:0', grad_fn=<AddBackward0>)
episode: 54 training return: tensor(-499.9439, device='cuda:0', grad_fn=<AddBackward0>)
episode: 55 training return: tensor(-499.8683, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 14 test_true_pfm: 1145.3221499075178
episode: 56 training return: tensor(-499.9715, device='cuda:0', grad_fn=<AddBackward0>)
episode: 57 training return: tensor(-499.9345, device='cuda:0', grad_fn=<AddBackward0>)
episode: 58 training return: tensor(-499.9685, device='cuda:0', grad_fn=<AddBackward0>)
episode: 59 training return: tensor(-499.9603, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 15 test_true_pfm: 707.6312650201693
episode: 60 training return: tensor(-499.8703, device='cuda:0', grad_fn=<AddBackward0>)
episode: 61 training return: tensor(-499.9672, device='cuda:0', grad_fn=<AddBackward0>)
episode: 62 training return: tensor(-499.9525, device='cuda:0', grad_fn=<AddBackward0>)
episode: 63 training return: tensor(-498.8170, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 16 test_true_pfm: 24.148187572549432
episode: 64 training return: tensor(-499.9569, device='cuda:0', grad_fn=<AddBackward0>)
episode: 65 training return: tensor(-499.9710, device='cuda:0', grad_fn=<AddBackward0>)
episode: 66 training return: tensor(-497.5143, device='cuda:0', grad_fn=<AddBackward0>)
episode: 67 training return: tensor(-499.9700, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 17 test_true_pfm: 844.53990221211
episode: 68 training return: tensor(-499.9592, device='cuda:0', grad_fn=<AddBackward0>)
episode: 69 training return: tensor(-499.6847, device='cuda:0', grad_fn=<AddBackward0>)
episode: 70 training return: tensor(-499.9709, device='cuda:0', grad_fn=<AddBackward0>)
episode: 71 training return: tensor(-498.9549, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 18 test_true_pfm: 1445.3515352818115
episode: 72 training return: tensor(-499.9691, device='cuda:0', grad_fn=<AddBackward0>)
episode: 73 training return: tensor(-499.9593, device='cuda:0', grad_fn=<AddBackward0>)
episode: 74 training return: tensor(-499.9553, device='cuda:0', grad_fn=<AddBackward0>)
episode: 75 training return: tensor(-499.9681, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 19 test_true_pfm: 1388.9610277152922
episode: 76 training return: tensor(-499.9503, device='cuda:0', grad_fn=<AddBackward0>)
episode: 77 training return: tensor(-499.9120, device='cuda:0', grad_fn=<AddBackward0>)
episode: 78 training return: tensor(-499.9704, device='cuda:0', grad_fn=<AddBackward0>)
episode: 79 training return: tensor(-494.8972, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 20 test_true_pfm: 2160.2360709787936
episode: 80 training return: tensor(-499.7853, device='cuda:0', grad_fn=<AddBackward0>)
episode: 81 training return: tensor(-460.6689, device='cuda:0', grad_fn=<AddBackward0>)
episode: 82 training return: tensor(-499.9592, device='cuda:0', grad_fn=<AddBackward0>)
episode: 83 training return: tensor(-429.7644, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 21 test_true_pfm: 593.2071437081237
episode: 84 training return: tensor(-452.9967, device='cuda:0', grad_fn=<AddBackward0>)
episode: 85 training return: tensor(-499.9002, device='cuda:0', grad_fn=<AddBackward0>)
episode: 86 training return: tensor(-449.3669, device='cuda:0', grad_fn=<AddBackward0>)
episode: 87 training return: tensor(-492.6586, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 22 test_true_pfm: 842.2201262879062
episode: 88 training return: tensor(-499.9678, device='cuda:0', grad_fn=<AddBackward0>)
episode: 89 training return: tensor(-497.1619, device='cuda:0', grad_fn=<AddBackward0>)
episode: 90 training return: tensor(-499.9536, device='cuda:0', grad_fn=<AddBackward0>)
episode: 91 training return: tensor(-499.9673, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 23 test_true_pfm: 3575.347871514145
episode: 92 training return: tensor(-497.1008, device='cuda:0', grad_fn=<AddBackward0>)
episode: 93 training return: tensor(-499.9661, device='cuda:0', grad_fn=<AddBackward0>)
episode: 94 training return: tensor(-499.9561, device='cuda:0', grad_fn=<AddBackward0>)
episode: 95 training return: tensor(-499.9722, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 24 test_true_pfm: 2197.964606068752
episode: 96 training return: tensor(-499.9556, device='cuda:0', grad_fn=<AddBackward0>)
episode: 97 training return: tensor(-499.9708, device='cuda:0', grad_fn=<AddBackward0>)
episode: 98 training return: tensor(-499.9714, device='cuda:0', grad_fn=<AddBackward0>)
episode: 99 training return: tensor(-499.9660, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 25 test_true_pfm: 987.4705354138799
episode: 100 training return: tensor(-499.9597, device='cuda:0', grad_fn=<AddBackward0>)
episode: 101 training return: tensor(-481.5391, device='cuda:0', grad_fn=<AddBackward0>)
episode: 102 training return: tensor(-499.9662, device='cuda:0', grad_fn=<AddBackward0>)
episode: 103 training return: tensor(-499.9693, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 26 test_true_pfm: -179.78048764409627
episode: 104 training return: tensor(-499.9679, device='cuda:0', grad_fn=<AddBackward0>)
episode: 105 training return: tensor(-499.9260, device='cuda:0', grad_fn=<AddBackward0>)
episode: 106 training return: tensor(-499.9702, device='cuda:0', grad_fn=<AddBackward0>)
episode: 107 training return: tensor(-496.3141, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 27 test_true_pfm: -237.44500731043806
episode: 108 training return: tensor(-499.9499, device='cuda:0', grad_fn=<AddBackward0>)
episode: 109 training return: tensor(-499.9431, device='cuda:0', grad_fn=<AddBackward0>)
episode: 110 training return: tensor(-499.9681, device='cuda:0', grad_fn=<AddBackward0>)
episode: 111 training return: tensor(-499.9353, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 28 test_true_pfm: -53.55246893739237
episode: 112 training return: tensor(-499.9655, device='cuda:0', grad_fn=<AddBackward0>)
episode: 113 training return: tensor(-499.9199, device='cuda:0', grad_fn=<AddBackward0>)
episode: 114 training return: tensor(-499.9623, device='cuda:0', grad_fn=<AddBackward0>)
episode: 115 training return: tensor(-471.9382, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 29 test_true_pfm: -216.33766413915797
episode: 116 training return: tensor(-499.9682, device='cuda:0', grad_fn=<AddBackward0>)
episode: 117 training return: tensor(-499.6816, device='cuda:0', grad_fn=<AddBackward0>)
episode: 118 training return: tensor(-499.9700, device='cuda:0', grad_fn=<AddBackward0>)
episode: 119 training return: tensor(-488.6979, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 30 test_true_pfm: 919.9013800958868
episode: 120 training return: tensor(-499.9215, device='cuda:0', grad_fn=<AddBackward0>)
episode: 121 training return: tensor(-497.3058, device='cuda:0', grad_fn=<AddBackward0>)
episode: 122 training return: tensor(-499.9693, device='cuda:0', grad_fn=<AddBackward0>)
episode: 123 training return: tensor(-499.9709, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 31 test_true_pfm: 2160.2865150452976
episode: 124 training return: tensor(-499.9281, device='cuda:0', grad_fn=<AddBackward0>)
episode: 125 training return: tensor(-499.8480, device='cuda:0', grad_fn=<AddBackward0>)
episode: 126 training return: tensor(-499.9699, device='cuda:0', grad_fn=<AddBackward0>)
episode: 127 training return: tensor(-499.9624, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 32 test_true_pfm: -362.57780923448627
episode: 128 training return: tensor(-495.5782, device='cuda:0', grad_fn=<AddBackward0>)
episode: 129 training return: tensor(-496.9439, device='cuda:0', grad_fn=<AddBackward0>)
episode: 130 training return: tensor(-499.9701, device='cuda:0', grad_fn=<AddBackward0>)
episode: 131 training return: tensor(-499.9631, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 33 test_true_pfm: 2122.9246663033027
episode: 132 training return: tensor(-499.9231, device='cuda:0', grad_fn=<AddBackward0>)
episode: 133 training return: tensor(-499.9691, device='cuda:0', grad_fn=<AddBackward0>)
episode: 134 training return: tensor(-499.9177, device='cuda:0', grad_fn=<AddBackward0>)
episode: 135 training return: tensor(-499.9518, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 34 test_true_pfm: 1065.9223406403116
episode: 136 training return: tensor(-499.9564, device='cuda:0', grad_fn=<AddBackward0>)
episode: 137 training return: tensor(-490.3412, device='cuda:0', grad_fn=<AddBackward0>)
episode: 138 training return: tensor(-499.5739, device='cuda:0', grad_fn=<AddBackward0>)
episode: 139 training return: tensor(-490.7519, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 35 test_true_pfm: 779.9234449720543
episode: 140 training return: tensor(-499.9670, device='cuda:0', grad_fn=<AddBackward0>)
episode: 141 training return: tensor(-499.9694, device='cuda:0', grad_fn=<AddBackward0>)
episode: 142 training return: tensor(-499.9705, device='cuda:0', grad_fn=<AddBackward0>)
episode: 143 training return: tensor(-499.9349, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 36 test_true_pfm: -314.9033953940529
episode: 144 training return: tensor(-499.9659, device='cuda:0', grad_fn=<AddBackward0>)
episode: 145 training return: tensor(-499.9608, device='cuda:0', grad_fn=<AddBackward0>)
episode: 146 training return: tensor(-499.9560, device='cuda:0', grad_fn=<AddBackward0>)
episode: 147 training return: tensor(-498.2022, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 37 test_true_pfm: -66.32406605021463
episode: 148 training return: tensor(-476.9860, device='cuda:0', grad_fn=<AddBackward0>)
episode: 149 training return: tensor(-499.9678, device='cuda:0', grad_fn=<AddBackward0>)
episode: 150 training return: tensor(-499.9700, device='cuda:0', grad_fn=<AddBackward0>)
episode: 151 training return: tensor(-499.9294, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 38 test_true_pfm: 1054.6789861286268
episode: 152 training return: tensor(-497.4208, device='cuda:0', grad_fn=<AddBackward0>)
episode: 153 training return: tensor(-499.4870, device='cuda:0', grad_fn=<AddBackward0>)
episode: 154 training return: tensor(-499.9709, device='cuda:0', grad_fn=<AddBackward0>)
episode: 155 training return: tensor(-499.9066, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 39 test_true_pfm: 1348.0141953055809
episode: 156 training return: tensor(-497.2795, device='cuda:0', grad_fn=<AddBackward0>)
episode: 157 training return: tensor(-499.9511, device='cuda:0', grad_fn=<AddBackward0>)
episode: 158 training return: tensor(-499.9403, device='cuda:0', grad_fn=<AddBackward0>)
episode: 159 training return: tensor(-499.9679, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 40 test_true_pfm: 4852.690073002758
episode: 160 training return: tensor(-499.9669, device='cuda:0', grad_fn=<AddBackward0>)
episode: 161 training return: tensor(-499.9587, device='cuda:0', grad_fn=<AddBackward0>)
episode: 162 training return: tensor(-482.3249, device='cuda:0', grad_fn=<AddBackward0>)
episode: 163 training return: tensor(-499.9255, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 41 test_true_pfm: 1165.2334402655076
episode: 164 training return: tensor(-499.6811, device='cuda:0', grad_fn=<AddBackward0>)
episode: 165 training return: tensor(-499.9125, device='cuda:0', grad_fn=<AddBackward0>)
episode: 166 training return: tensor(-499.9548, device='cuda:0', grad_fn=<AddBackward0>)
episode: 167 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 42 test_true_pfm: -248.91823130135617
episode: 168 training return: tensor(-499.8926, device='cuda:0', grad_fn=<AddBackward0>)
episode: 169 training return: tensor(-499.8407, device='cuda:0', grad_fn=<AddBackward0>)
episode: 170 training return: tensor(-491.0716, device='cuda:0', grad_fn=<AddBackward0>)
episode: 171 training return: tensor(-499.9701, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 43 test_true_pfm: 942.8698244987909
episode: 172 training return: tensor(-499.8477, device='cuda:0', grad_fn=<AddBackward0>)
episode: 173 training return: tensor(-499.9702, device='cuda:0', grad_fn=<AddBackward0>)
episode: 174 training return: tensor(-499.9231, device='cuda:0', grad_fn=<AddBackward0>)
episode: 175 training return: tensor(-499.9221, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 44 test_true_pfm: 741.0457726962644
episode: 176 training return: tensor(-496.9196, device='cuda:0', grad_fn=<AddBackward0>)
episode: 177 training return: tensor(-493.1620, device='cuda:0', grad_fn=<AddBackward0>)
episode: 178 training return: tensor(-498.8555, device='cuda:0', grad_fn=<AddBackward0>)
episode: 179 training return: tensor(-496.0606, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 45 test_true_pfm: -174.59323099505409
episode: 180 training return: tensor(-499.9465, device='cuda:0', grad_fn=<AddBackward0>)
episode: 181 training return: tensor(-499.9663, device='cuda:0', grad_fn=<AddBackward0>)
episode: 182 training return: tensor(-499.9586, device='cuda:0', grad_fn=<AddBackward0>)
episode: 183 training return: tensor(-499.9584, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 46 test_true_pfm: 1573.9697163727699
episode: 184 training return: tensor(-499.9076, device='cuda:0', grad_fn=<AddBackward0>)
episode: 185 training return: tensor(-492.1510, device='cuda:0', grad_fn=<AddBackward0>)
episode: 186 training return: tensor(-470.6249, device='cuda:0', grad_fn=<AddBackward0>)
episode: 187 training return: tensor(-498.7234, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 47 test_true_pfm: 1557.7036902901345
episode: 188 training return: tensor(-499.9563, device='cuda:0', grad_fn=<AddBackward0>)
episode: 189 training return: tensor(-477.7708, device='cuda:0', grad_fn=<AddBackward0>)
episode: 190 training return: tensor(-499.9706, device='cuda:0', grad_fn=<AddBackward0>)
episode: 191 training return: tensor(-499.9696, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 48 test_true_pfm: 901.9220892603513
episode: 192 training return: tensor(-499.7164, device='cuda:0', grad_fn=<AddBackward0>)
episode: 193 training return: tensor(-499.9600, device='cuda:0', grad_fn=<AddBackward0>)
episode: 194 training return: tensor(-499.9388, device='cuda:0', grad_fn=<AddBackward0>)
episode: 195 training return: tensor(-425.5898, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 49 test_true_pfm: 816.0103023925668
episode: 196 training return: tensor(-499.8264, device='cuda:0', grad_fn=<AddBackward0>)
episode: 197 training return: tensor(-499.9662, device='cuda:0', grad_fn=<AddBackward0>)
episode: 198 training return: tensor(-499.8273, device='cuda:0', grad_fn=<AddBackward0>)
episode: 199 training return: tensor(-499.9686, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 50 test_true_pfm: 1224.0026393202509
episode: 200 training return: tensor(-499.9666, device='cuda:0', grad_fn=<AddBackward0>)
episode: 201 training return: tensor(-499.9678, device='cuda:0', grad_fn=<AddBackward0>)
episode: 202 training return: tensor(-499.8351, device='cuda:0', grad_fn=<AddBackward0>)
episode: 203 training return: tensor(-499.9074, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 51 test_true_pfm: 753.4510368829382
episode: 204 training return: tensor(-499.9456, device='cuda:0', grad_fn=<AddBackward0>)
episode: 205 training return: tensor(-499.9710, device='cuda:0', grad_fn=<AddBackward0>)
episode: 206 training return: tensor(-499.9441, device='cuda:0', grad_fn=<AddBackward0>)
episode: 207 training return: tensor(-489.1419, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 52 test_true_pfm: 1300.7498444645287
episode: 208 training return: tensor(-499.9706, device='cuda:0', grad_fn=<AddBackward0>)
episode: 209 training return: tensor(-499.7394, device='cuda:0', grad_fn=<AddBackward0>)
episode: 210 training return: tensor(-499.9673, device='cuda:0', grad_fn=<AddBackward0>)
episode: 211 training return: tensor(-499.9597, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 53 test_true_pfm: -70.08818433757237
episode: 212 training return: tensor(-499.9489, device='cuda:0', grad_fn=<AddBackward0>)
episode: 213 training return: tensor(-499.9690, device='cuda:0', grad_fn=<AddBackward0>)
episode: 214 training return: tensor(-499.7247, device='cuda:0', grad_fn=<AddBackward0>)
episode: 215 training return: tensor(-499.9398, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 54 test_true_pfm: 2500.3992799847106
episode: 216 training return: tensor(-499.9595, device='cuda:0', grad_fn=<AddBackward0>)
episode: 217 training return: tensor(-499.9407, device='cuda:0', grad_fn=<AddBackward0>)
episode: 218 training return: tensor(-499.9569, device='cuda:0', grad_fn=<AddBackward0>)
episode: 219 training return: tensor(-499.9692, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 55 test_true_pfm: 595.4615786097612
episode: 220 training return: tensor(-499.9693, device='cuda:0', grad_fn=<AddBackward0>)
episode: 221 training return: tensor(-499.9711, device='cuda:0', grad_fn=<AddBackward0>)
episode: 222 training return: tensor(-499.9678, device='cuda:0', grad_fn=<AddBackward0>)
episode: 223 training return: tensor(-499.9487, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 56 test_true_pfm: -282.92054818824164
episode: 224 training return: tensor(-499.9638, device='cuda:0', grad_fn=<AddBackward0>)
episode: 225 training return: tensor(-499.9388, device='cuda:0', grad_fn=<AddBackward0>)
episode: 226 training return: tensor(-499.9431, device='cuda:0', grad_fn=<AddBackward0>)
episode: 227 training return: tensor(-494.7669, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 57 test_true_pfm: 757.7885967426113
episode: 228 training return: tensor(-499.3624, device='cuda:0', grad_fn=<AddBackward0>)
episode: 229 training return: tensor(-493.3578, device='cuda:0', grad_fn=<AddBackward0>)
episode: 230 training return: tensor(-499.9545, device='cuda:0', grad_fn=<AddBackward0>)
episode: 231 training return: tensor(-499.1536, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 58 test_true_pfm: 872.1616814090512
episode: 232 training return: tensor(-499.9684, device='cuda:0', grad_fn=<AddBackward0>)
episode: 233 training return: tensor(-486.7817, device='cuda:0', grad_fn=<AddBackward0>)
episode: 234 training return: tensor(-499.9693, device='cuda:0', grad_fn=<AddBackward0>)
episode: 235 training return: tensor(-499.9095, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 59 test_true_pfm: -174.21558447874813
episode: 236 training return: tensor(-499.9558, device='cuda:0', grad_fn=<AddBackward0>)
episode: 237 training return: tensor(-499.9257, device='cuda:0', grad_fn=<AddBackward0>)
episode: 238 training return: tensor(-499.8513, device='cuda:0', grad_fn=<AddBackward0>)
episode: 239 training return: tensor(-499.9638, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 60 test_true_pfm: 5118.218617392071
episode: 240 training return: tensor(-499.6496, device='cuda:0', grad_fn=<AddBackward0>)
episode: 241 training return: tensor(-499.9402, device='cuda:0', grad_fn=<AddBackward0>)
episode: 242 training return: tensor(-499.9654, device='cuda:0', grad_fn=<AddBackward0>)
episode: 243 training return: tensor(-499.9682, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 61 test_true_pfm: 3973.7105075291893
episode: 244 training return: tensor(-499.9363, device='cuda:0', grad_fn=<AddBackward0>)
episode: 245 training return: tensor(-499.9457, device='cuda:0', grad_fn=<AddBackward0>)
episode: 246 training return: tensor(-458.1287, device='cuda:0', grad_fn=<AddBackward0>)
episode: 247 training return: tensor(-499.9116, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 62 test_true_pfm: 166.80002467934028
episode: 248 training return: tensor(-499.9669, device='cuda:0', grad_fn=<AddBackward0>)
episode: 249 training return: tensor(-486.2974, device='cuda:0', grad_fn=<AddBackward0>)
episode: 250 training return: tensor(-446.0251, device='cuda:0', grad_fn=<AddBackward0>)
episode: 251 training return: tensor(-499.9605, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 63 test_true_pfm: 3678.240846239542
episode: 252 training return: tensor(-499.5894, device='cuda:0', grad_fn=<AddBackward0>)
episode: 253 training return: tensor(-496.5422, device='cuda:0', grad_fn=<AddBackward0>)
episode: 254 training return: tensor(-499.9655, device='cuda:0', grad_fn=<AddBackward0>)
episode: 255 training return: tensor(-499.9590, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 64 test_true_pfm: 842.7233488820139
episode: 256 training return: tensor(-499.9611, device='cuda:0', grad_fn=<AddBackward0>)
episode: 257 training return: tensor(-499.9699, device='cuda:0', grad_fn=<AddBackward0>)
episode: 258 training return: tensor(-499.9505, device='cuda:0', grad_fn=<AddBackward0>)
episode: 259 training return: tensor(-487.2694, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 65 test_true_pfm: 879.3738902250149
episode: 260 training return: tensor(-499.9705, device='cuda:0', grad_fn=<AddBackward0>)
episode: 261 training return: tensor(-499.1597, device='cuda:0', grad_fn=<AddBackward0>)
episode: 262 training return: tensor(-498.5912, device='cuda:0', grad_fn=<AddBackward0>)
episode: 263 training return: tensor(-499.9685, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 66 test_true_pfm: -391.7827982695526
episode: 264 training return: tensor(-498.3565, device='cuda:0', grad_fn=<AddBackward0>)
episode: 265 training return: tensor(-499.9695, device='cuda:0', grad_fn=<AddBackward0>)
episode: 266 training return: tensor(-499.9583, device='cuda:0', grad_fn=<AddBackward0>)
episode: 267 training return: tensor(-499.4709, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 67 test_true_pfm: 331.7769297169886
episode: 268 training return: tensor(-448.0577, device='cuda:0', grad_fn=<AddBackward0>)
episode: 269 training return: tensor(-499.9666, device='cuda:0', grad_fn=<AddBackward0>)
episode: 270 training return: tensor(-499.9646, device='cuda:0', grad_fn=<AddBackward0>)
episode: 271 training return: tensor(-499.9435, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 68 test_true_pfm: 3892.646205739546
episode: 272 training return: tensor(-499.9424, device='cuda:0', grad_fn=<AddBackward0>)
episode: 273 training return: tensor(-499.7241, device='cuda:0', grad_fn=<AddBackward0>)
episode: 274 training return: tensor(-499.0038, device='cuda:0', grad_fn=<AddBackward0>)
episode: 275 training return: tensor(-499.8729, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 69 test_true_pfm: 776.5229339200972
episode: 276 training return: tensor(-499.9492, device='cuda:0', grad_fn=<AddBackward0>)
episode: 277 training return: tensor(-499.6765, device='cuda:0', grad_fn=<AddBackward0>)
episode: 278 training return: tensor(-493.7011, device='cuda:0', grad_fn=<AddBackward0>)
episode: 279 training return: tensor(-499.7837, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 70 test_true_pfm: 3550.9996335740966
episode: 280 training return: tensor(-499.9378, device='cuda:0', grad_fn=<AddBackward0>)
episode: 281 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
episode: 282 training return: tensor(-493.2562, device='cuda:0', grad_fn=<AddBackward0>)
episode: 283 training return: tensor(-499.9489, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 71 test_true_pfm: 203.2003843088415
episode: 284 training return: tensor(-497.9438, device='cuda:0', grad_fn=<AddBackward0>)
episode: 285 training return: tensor(-499.9585, device='cuda:0', grad_fn=<AddBackward0>)
episode: 286 training return: tensor(-499.8951, device='cuda:0', grad_fn=<AddBackward0>)
episode: 287 training return: tensor(-499.9699, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 72 test_true_pfm: -262.91596769351094
episode: 288 training return: tensor(-499.9591, device='cuda:0', grad_fn=<AddBackward0>)
episode: 289 training return: tensor(-497.8744, device='cuda:0', grad_fn=<AddBackward0>)
episode: 290 training return: tensor(-497.5525, device='cuda:0', grad_fn=<AddBackward0>)
episode: 291 training return: tensor(-496.1141, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 73 test_true_pfm: -191.27376926382263
episode: 292 training return: tensor(-499.9464, device='cuda:0', grad_fn=<AddBackward0>)
episode: 293 training return: tensor(-499.4320, device='cuda:0', grad_fn=<AddBackward0>)
episode: 294 training return: tensor(-499.9694, device='cuda:0', grad_fn=<AddBackward0>)
episode: 295 training return: tensor(-499.1837, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 74 test_true_pfm: 187.3834534249916
episode: 296 training return: tensor(-499.9663, device='cuda:0', grad_fn=<AddBackward0>)
episode: 297 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
episode: 298 training return: tensor(-499.8455, device='cuda:0', grad_fn=<AddBackward0>)
episode: 299 training return: tensor(-499.9690, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 75 test_true_pfm: 578.0693966239395
episode: 300 training return: tensor(-499.9623, device='cuda:0', grad_fn=<AddBackward0>)
episode: 301 training return: tensor(-499.9232, device='cuda:0', grad_fn=<AddBackward0>)
episode: 302 training return: tensor(-497.3986, device='cuda:0', grad_fn=<AddBackward0>)
episode: 303 training return: tensor(-497.1269, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 76 test_true_pfm: 1139.8985960900593
episode: 304 training return: tensor(-499.9567, device='cuda:0', grad_fn=<AddBackward0>)
episode: 305 training return: tensor(-499.9203, device='cuda:0', grad_fn=<AddBackward0>)
episode: 306 training return: tensor(-499.9665, device='cuda:0', grad_fn=<AddBackward0>)
episode: 307 training return: tensor(-499.9688, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 77 test_true_pfm: 2809.4881284309195
episode: 308 training return: tensor(-499.9606, device='cuda:0', grad_fn=<AddBackward0>)
episode: 309 training return: tensor(-498.6714, device='cuda:0', grad_fn=<AddBackward0>)
episode: 310 training return: tensor(-499.9697, device='cuda:0', grad_fn=<AddBackward0>)
episode: 311 training return: tensor(-499.9695, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 78 test_true_pfm: 821.2874465372528
episode: 312 training return: tensor(-499.9223, device='cuda:0', grad_fn=<AddBackward0>)
episode: 313 training return: tensor(-499.9698, device='cuda:0', grad_fn=<AddBackward0>)
episode: 314 training return: tensor(-499.9436, device='cuda:0', grad_fn=<AddBackward0>)
episode: 315 training return: tensor(-445.9674, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 79 test_true_pfm: 1476.2922098832985
episode: 316 training return: tensor(-499.9387, device='cuda:0', grad_fn=<AddBackward0>)
episode: 317 training return: tensor(-499.9646, device='cuda:0', grad_fn=<AddBackward0>)
episode: 318 training return: tensor(-499.9638, device='cuda:0', grad_fn=<AddBackward0>)
episode: 319 training return: tensor(-499.9603, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 80 test_true_pfm: -303.98712639671004
episode: 320 training return: tensor(-499.9666, device='cuda:0', grad_fn=<AddBackward0>)
episode: 321 training return: tensor(-499.9576, device='cuda:0', grad_fn=<AddBackward0>)
episode: 322 training return: tensor(-499.9698, device='cuda:0', grad_fn=<AddBackward0>)
episode: 323 training return: tensor(-499.9077, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 81 test_true_pfm: -64.48161545224555
episode: 324 training return: tensor(-499.9389, device='cuda:0', grad_fn=<AddBackward0>)
episode: 325 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
episode: 326 training return: tensor(-499.9669, device='cuda:0', grad_fn=<AddBackward0>)
episode: 327 training return: tensor(-499.9189, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 82 test_true_pfm: 3780.2001464311966
episode: 328 training return: tensor(-499.9700, device='cuda:0', grad_fn=<AddBackward0>)
episode: 329 training return: tensor(-499.9596, device='cuda:0', grad_fn=<AddBackward0>)
episode: 330 training return: tensor(-499.9696, device='cuda:0', grad_fn=<AddBackward0>)
episode: 331 training return: tensor(-485.3651, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 83 test_true_pfm: -241.1147720581283
episode: 332 training return: tensor(-499.9587, device='cuda:0', grad_fn=<AddBackward0>)
episode: 333 training return: tensor(-495.2987, device='cuda:0', grad_fn=<AddBackward0>)
episode: 334 training return: tensor(-499.8356, device='cuda:0', grad_fn=<AddBackward0>)
episode: 335 training return: tensor(-499.9683, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 84 test_true_pfm: 1122.1719936474678
episode: 336 training return: tensor(-499.9213, device='cuda:0', grad_fn=<AddBackward0>)
episode: 337 training return: tensor(-499.9690, device='cuda:0', grad_fn=<AddBackward0>)
episode: 338 training return: tensor(-499.9268, device='cuda:0', grad_fn=<AddBackward0>)
episode: 339 training return: tensor(-499.9415, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 85 test_true_pfm: -285.93825631362853
episode: 340 training return: tensor(-494.4513, device='cuda:0', grad_fn=<AddBackward0>)
episode: 341 training return: tensor(-499.9684, device='cuda:0', grad_fn=<AddBackward0>)
episode: 342 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
episode: 343 training return: tensor(-498.4846, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 86 test_true_pfm: 1434.1877546106728
episode: 344 training return: tensor(-499.9427, device='cuda:0', grad_fn=<AddBackward0>)
episode: 345 training return: tensor(-495.4244, device='cuda:0', grad_fn=<AddBackward0>)
episode: 346 training return: tensor(-499.9706, device='cuda:0', grad_fn=<AddBackward0>)
episode: 347 training return: tensor(-499.9701, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 87 test_true_pfm: 2075.986307327678
episode: 348 training return: tensor(-478.1073, device='cuda:0', grad_fn=<AddBackward0>)
episode: 349 training return: tensor(-499.9697, device='cuda:0', grad_fn=<AddBackward0>)
episode: 350 training return: tensor(-499.9689, device='cuda:0', grad_fn=<AddBackward0>)
episode: 351 training return: tensor(-485.1272, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 88 test_true_pfm: 3220.050942489382
episode: 352 training return: tensor(-499.9441, device='cuda:0', grad_fn=<AddBackward0>)
episode: 353 training return: tensor(-499.9586, device='cuda:0', grad_fn=<AddBackward0>)
episode: 354 training return: tensor(-499.9708, device='cuda:0', grad_fn=<AddBackward0>)
episode: 355 training return: tensor(-499.9531, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 89 test_true_pfm: 1566.3925362528096
episode: 356 training return: tensor(-499.9363, device='cuda:0', grad_fn=<AddBackward0>)
episode: 357 training return: tensor(-499.7711, device='cuda:0', grad_fn=<AddBackward0>)
episode: 358 training return: tensor(-499.9507, device='cuda:0', grad_fn=<AddBackward0>)
episode: 359 training return: tensor(-499.8889, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 90 test_true_pfm: 1224.2941092587826
episode: 360 training return: tensor(-499.9110, device='cuda:0', grad_fn=<AddBackward0>)
episode: 361 training return: tensor(-499.8080, device='cuda:0', grad_fn=<AddBackward0>)
episode: 362 training return: tensor(-499.9642, device='cuda:0', grad_fn=<AddBackward0>)
episode: 363 training return: tensor(-497.0394, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 91 test_true_pfm: 427.6751475356184
episode: 364 training return: tensor(-499.9645, device='cuda:0', grad_fn=<AddBackward0>)
episode: 365 training return: tensor(-498.7100, device='cuda:0', grad_fn=<AddBackward0>)
episode: 366 training return: tensor(-499.9408, device='cuda:0', grad_fn=<AddBackward0>)
episode: 367 training return: tensor(-499.9491, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 92 test_true_pfm: 547.248337488774
episode: 368 training return: tensor(-499.9623, device='cuda:0', grad_fn=<AddBackward0>)
episode: 369 training return: tensor(-499.9668, device='cuda:0', grad_fn=<AddBackward0>)
episode: 370 training return: tensor(-499.9585, device='cuda:0', grad_fn=<AddBackward0>)
episode: 371 training return: tensor(-499.9589, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 93 test_true_pfm: 1541.0174917142967
episode: 372 training return: tensor(-499.1256, device='cuda:0', grad_fn=<AddBackward0>)
episode: 373 training return: tensor(-499.9692, device='cuda:0', grad_fn=<AddBackward0>)
episode: 374 training return: tensor(-499.9523, device='cuda:0', grad_fn=<AddBackward0>)
episode: 375 training return: tensor(-499.9456, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 94 test_true_pfm: 1073.41177714548
episode: 376 training return: tensor(-499.9691, device='cuda:0', grad_fn=<AddBackward0>)
episode: 377 training return: tensor(-493.6781, device='cuda:0', grad_fn=<AddBackward0>)
episode: 378 training return: tensor(-499.9670, device='cuda:0', grad_fn=<AddBackward0>)
episode: 379 training return: tensor(-499.9544, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 95 test_true_pfm: -54.260408308145834
episode: 380 training return: tensor(-499.9356, device='cuda:0', grad_fn=<AddBackward0>)
episode: 381 training return: tensor(-499.9709, device='cuda:0', grad_fn=<AddBackward0>)
episode: 382 training return: tensor(-499.9282, device='cuda:0', grad_fn=<AddBackward0>)
episode: 383 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 96 test_true_pfm: 400.89221139085606
episode: 384 training return: tensor(-499.9664, device='cuda:0', grad_fn=<AddBackward0>)
episode: 385 training return: tensor(-499.3561, device='cuda:0', grad_fn=<AddBackward0>)
episode: 386 training return: tensor(-499.9378, device='cuda:0', grad_fn=<AddBackward0>)
episode: 387 training return: tensor(-499.9715, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 97 test_true_pfm: 1580.064763615601
episode: 388 training return: tensor(-499.9677, device='cuda:0', grad_fn=<AddBackward0>)
episode: 389 training return: tensor(-499.9699, device='cuda:0', grad_fn=<AddBackward0>)
episode: 390 training return: tensor(-486.0753, device='cuda:0', grad_fn=<AddBackward0>)
episode: 391 training return: tensor(-499.9376, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 98 test_true_pfm: 402.54746327017284
episode: 392 training return: tensor(-499.9403, device='cuda:0', grad_fn=<AddBackward0>)
episode: 393 training return: tensor(-499.9516, device='cuda:0', grad_fn=<AddBackward0>)
episode: 394 training return: tensor(-499.9701, device='cuda:0', grad_fn=<AddBackward0>)
episode: 395 training return: tensor(-498.5118, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 99 test_true_pfm: -53.974718256301834
episode: 396 training return: tensor(-483.7198, device='cuda:0', grad_fn=<AddBackward0>)
episode: 397 training return: tensor(-499.9404, device='cuda:0', grad_fn=<AddBackward0>)
episode: 398 training return: tensor(-476.9120, device='cuda:0', grad_fn=<AddBackward0>)
episode: 399 training return: tensor(-499.8804, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 100 test_true_pfm: 1816.2378672879968
episode: 400 training return: tensor(-499.9616, device='cuda:0', grad_fn=<AddBackward0>)
episode: 401 training return: tensor(-499.9440, device='cuda:0', grad_fn=<AddBackward0>)
episode: 402 training return: tensor(-499.9650, device='cuda:0', grad_fn=<AddBackward0>)
episode: 403 training return: tensor(-498.3936, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 101 test_true_pfm: 2976.639070041431
episode: 404 training return: tensor(-499.9699, device='cuda:0', grad_fn=<AddBackward0>)
episode: 405 training return: tensor(-499.9633, device='cuda:0', grad_fn=<AddBackward0>)
episode: 406 training return: tensor(-499.3804, device='cuda:0', grad_fn=<AddBackward0>)
episode: 407 training return: tensor(-499.9288, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 102 test_true_pfm: 76.53802878428526
episode: 408 training return: tensor(-499.9467, device='cuda:0', grad_fn=<AddBackward0>)
episode: 409 training return: tensor(-484.8489, device='cuda:0', grad_fn=<AddBackward0>)
episode: 410 training return: tensor(-499.9649, device='cuda:0', grad_fn=<AddBackward0>)
episode: 411 training return: tensor(-499.9700, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 103 test_true_pfm: 2217.345097183433
episode: 412 training return: tensor(-499.6657, device='cuda:0', grad_fn=<AddBackward0>)
episode: 413 training return: tensor(-499.9673, device='cuda:0', grad_fn=<AddBackward0>)
episode: 414 training return: tensor(-499.9679, device='cuda:0', grad_fn=<AddBackward0>)
episode: 415 training return: tensor(-499.9236, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 104 test_true_pfm: 802.4151012576082
episode: 416 training return: tensor(-499.9656, device='cuda:0', grad_fn=<AddBackward0>)
episode: 417 training return: tensor(-499.2810, device='cuda:0', grad_fn=<AddBackward0>)
episode: 418 training return: tensor(-499.9482, device='cuda:0', grad_fn=<AddBackward0>)
episode: 419 training return: tensor(-499.9703, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 105 test_true_pfm: 168.30295334275525
episode: 420 training return: tensor(-496.7160, device='cuda:0', grad_fn=<AddBackward0>)
episode: 421 training return: tensor(-499.9471, device='cuda:0', grad_fn=<AddBackward0>)
episode: 422 training return: tensor(-457.6195, device='cuda:0', grad_fn=<AddBackward0>)
episode: 423 training return: tensor(-499.8705, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 106 test_true_pfm: 1036.1588862678802
episode: 424 training return: tensor(-499.9691, device='cuda:0', grad_fn=<AddBackward0>)
episode: 425 training return: tensor(-499.9443, device='cuda:0', grad_fn=<AddBackward0>)
episode: 426 training return: tensor(-499.9191, device='cuda:0', grad_fn=<AddBackward0>)
episode: 427 training return: tensor(-499.9145, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 107 test_true_pfm: 2036.7060655315934
episode: 428 training return: tensor(-490.7940, device='cuda:0', grad_fn=<AddBackward0>)
episode: 429 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
episode: 430 training return: tensor(-499.9698, device='cuda:0', grad_fn=<AddBackward0>)
episode: 431 training return: tensor(-499.9466, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 108 test_true_pfm: 1054.634410725346
episode: 432 training return: tensor(-499.9511, device='cuda:0', grad_fn=<AddBackward0>)
episode: 433 training return: tensor(-499.9530, device='cuda:0', grad_fn=<AddBackward0>)
episode: 434 training return: tensor(-499.9697, device='cuda:0', grad_fn=<AddBackward0>)
episode: 435 training return: tensor(-499.9694, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 109 test_true_pfm: 630.4585582419198
episode: 436 training return: tensor(-499.9681, device='cuda:0', grad_fn=<AddBackward0>)
episode: 437 training return: tensor(-499.9691, device='cuda:0', grad_fn=<AddBackward0>)
episode: 438 training return: tensor(-499.9418, device='cuda:0', grad_fn=<AddBackward0>)
episode: 439 training return: tensor(-497.8057, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 110 test_true_pfm: 3111.801750268791
episode: 440 training return: tensor(-499.9687, device='cuda:0', grad_fn=<AddBackward0>)
episode: 441 training return: tensor(-499.9445, device='cuda:0', grad_fn=<AddBackward0>)
episode: 442 training return: tensor(-499.9237, device='cuda:0', grad_fn=<AddBackward0>)
episode: 443 training return: tensor(-499.9305, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 111 test_true_pfm: -72.75695085293285
episode: 444 training return: tensor(-499.9661, device='cuda:0', grad_fn=<AddBackward0>)
episode: 445 training return: tensor(-499.9658, device='cuda:0', grad_fn=<AddBackward0>)
episode: 446 training return: tensor(-492.7285, device='cuda:0', grad_fn=<AddBackward0>)
episode: 447 training return: tensor(-498.6476, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 112 test_true_pfm: 2590.315634106566
episode: 448 training return: tensor(-499.9659, device='cuda:0', grad_fn=<AddBackward0>)
episode: 449 training return: tensor(-499.9593, device='cuda:0', grad_fn=<AddBackward0>)
episode: 450 training return: tensor(-499.9688, device='cuda:0', grad_fn=<AddBackward0>)
episode: 451 training return: tensor(-499.9263, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 113 test_true_pfm: 887.9139500613132
episode: 452 training return: tensor(-499.9683, device='cuda:0', grad_fn=<AddBackward0>)
episode: 453 training return: tensor(-499.9698, device='cuda:0', grad_fn=<AddBackward0>)
episode: 454 training return: tensor(-497.4862, device='cuda:0', grad_fn=<AddBackward0>)
episode: 455 training return: tensor(-499.9706, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 114 test_true_pfm: 2654.815255576042
episode: 456 training return: tensor(-499.9648, device='cuda:0', grad_fn=<AddBackward0>)
episode: 457 training return: tensor(-499.9570, device='cuda:0', grad_fn=<AddBackward0>)
episode: 458 training return: tensor(-499.6879, device='cuda:0', grad_fn=<AddBackward0>)
episode: 459 training return: tensor(-499.9692, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 115 test_true_pfm: 1459.9561089594145
episode: 460 training return: tensor(-496.4869, device='cuda:0', grad_fn=<AddBackward0>)
episode: 461 training return: tensor(-499.9691, device='cuda:0', grad_fn=<AddBackward0>)
episode: 462 training return: tensor(-499.9691, device='cuda:0', grad_fn=<AddBackward0>)
episode: 463 training return: tensor(-499.5051, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 116 test_true_pfm: 888.7825472609314
episode: 464 training return: tensor(-499.9394, device='cuda:0', grad_fn=<AddBackward0>)
episode: 465 training return: tensor(-499.9186, device='cuda:0', grad_fn=<AddBackward0>)
episode: 466 training return: tensor(-499.9705, device='cuda:0', grad_fn=<AddBackward0>)
episode: 467 training return: tensor(-499.9519, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 117 test_true_pfm: 978.4794568384431
episode: 468 training return: tensor(-499.9634, device='cuda:0', grad_fn=<AddBackward0>)
episode: 469 training return: tensor(-485.2266, device='cuda:0', grad_fn=<AddBackward0>)
episode: 470 training return: tensor(-499.9425, device='cuda:0', grad_fn=<AddBackward0>)
episode: 471 training return: tensor(-499.9697, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 118 test_true_pfm: 2277.1142543899136
episode: 472 training return: tensor(-499.9655, device='cuda:0', grad_fn=<AddBackward0>)
episode: 473 training return: tensor(-499.9430, device='cuda:0', grad_fn=<AddBackward0>)
episode: 474 training return: tensor(-499.9603, device='cuda:0', grad_fn=<AddBackward0>)
episode: 475 training return: tensor(-498.8120, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 119 test_true_pfm: 115.00820695599948
episode: 476 training return: tensor(-499.9644, device='cuda:0', grad_fn=<AddBackward0>)
episode: 477 training return: tensor(-499.9544, device='cuda:0', grad_fn=<AddBackward0>)
episode: 478 training return: tensor(-453.1425, device='cuda:0', grad_fn=<AddBackward0>)
episode: 479 training return: tensor(-499.9612, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 120 test_true_pfm: -8.376431924773746
episode: 480 training return: tensor(-493.8458, device='cuda:0', grad_fn=<AddBackward0>)
episode: 481 training return: tensor(-499.9436, device='cuda:0', grad_fn=<AddBackward0>)
episode: 482 training return: tensor(-499.7381, device='cuda:0', grad_fn=<AddBackward0>)
episode: 483 training return: tensor(-499.9050, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 121 test_true_pfm: -270.23318023617037
episode: 484 training return: tensor(-499.9679, device='cuda:0', grad_fn=<AddBackward0>)
episode: 485 training return: tensor(-499.9331, device='cuda:0', grad_fn=<AddBackward0>)
episode: 486 training return: tensor(-499.9671, device='cuda:0', grad_fn=<AddBackward0>)
episode: 487 training return: tensor(-499.9637, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 122 test_true_pfm: 2922.7090779795885
episode: 488 training return: tensor(-499.7912, device='cuda:0', grad_fn=<AddBackward0>)
episode: 489 training return: tensor(-494.6910, device='cuda:0', grad_fn=<AddBackward0>)
episode: 490 training return: tensor(-499.9445, device='cuda:0', grad_fn=<AddBackward0>)
episode: 491 training return: tensor(-499.9698, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 123 test_true_pfm: 698.7899841788008
episode: 492 training return: tensor(-499.9698, device='cuda:0', grad_fn=<AddBackward0>)
episode: 493 training return: tensor(-490.9532, device='cuda:0', grad_fn=<AddBackward0>)
episode: 494 training return: tensor(-499.8204, device='cuda:0', grad_fn=<AddBackward0>)
episode: 495 training return: tensor(-498.6205, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 124 test_true_pfm: 1636.1557450236871
episode: 496 training return: tensor(-499.9351, device='cuda:0', grad_fn=<AddBackward0>)
episode: 497 training return: tensor(-499.9649, device='cuda:0', grad_fn=<AddBackward0>)
episode: 498 training return: tensor(-499.8730, device='cuda:0', grad_fn=<AddBackward0>)
episode: 499 training return: tensor(-499.9693, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 125 test_true_pfm: 2887.4651382989027
episode: 500 training return: tensor(-499.9704, device='cuda:0', grad_fn=<AddBackward0>)
episode: 501 training return: tensor(-499.9134, device='cuda:0', grad_fn=<AddBackward0>)
episode: 502 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
episode: 503 training return: tensor(-499.9706, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 126 test_true_pfm: 1359.8688686460491
episode: 504 training return: tensor(-499.9662, device='cuda:0', grad_fn=<AddBackward0>)
episode: 505 training return: tensor(-499.9713, device='cuda:0', grad_fn=<AddBackward0>)
episode: 506 training return: tensor(-499.9714, device='cuda:0', grad_fn=<AddBackward0>)
episode: 507 training return: tensor(-499.9306, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 127 test_true_pfm: 2996.630563835612
episode: 508 training return: tensor(-498.9495, device='cuda:0', grad_fn=<AddBackward0>)
episode: 509 training return: tensor(-499.9688, device='cuda:0', grad_fn=<AddBackward0>)
episode: 510 training return: tensor(-497.0865, device='cuda:0', grad_fn=<AddBackward0>)
episode: 511 training return: tensor(-499.9696, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 128 test_true_pfm: -200.30138987579127
episode: 512 training return: tensor(-499.9473, device='cuda:0', grad_fn=<AddBackward0>)
episode: 513 training return: tensor(-499.9697, device='cuda:0', grad_fn=<AddBackward0>)
episode: 514 training return: tensor(-499.9659, device='cuda:0', grad_fn=<AddBackward0>)
episode: 515 training return: tensor(-499.8595, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 129 test_true_pfm: 111.72667335077246
episode: 516 training return: tensor(-499.9693, device='cuda:0', grad_fn=<AddBackward0>)
